Introduction
Multiple infections occur whenever different pathogen species or strains affect the same host (reviewed by Schmid-Hempel, Reference Schmid-Hempel2011; Mahmud et al., Reference Mahmud, Lim and Amir2018). Compared with infections by a single pathogen, multiple infections have different evolutionary effects, like changes in virulence (defined as the damage by the parasite on the host; Read, Reference Read1994) of the pathogens involved (López-Villavicencio et al., Reference López-Villavicencio, Courjol, Gibson, Hood, Jonot, Shykoff and Giraud2011; Alizon et al., Reference Alizon, de Roode and Michalakis2013), in the resources obtained by the pathogens (e.g. Choisy and Roode, Reference Choisy and de Roode2010), and in host fitness (Bordes and Morand, Reference Bordes and Morand2011; Kamiya et al., Reference Kamiya, Mideo and Alizon2018). Understanding that most pathogen–host interactions involve multiple infections has implications in a variety of topics, including the dynamics of diseases transmitted by pest insects and vectors, and their control.
While multiple infections have been used in pest biological control (reviewed by Rechcigl and Rechcigl, Reference Rechcigl and Rechcigl2016; Poinar, Reference Poinar2018), this approach has not been widely used in vectors (Benelli, Reference Benelli2018). In this regard, most studies have been conducted on mosquito species responsible for spreading viral diseases (Lorono-Pino et al., Reference Lorono-Pino, Cropp, Farfan, Vorndam, Rodríguez-Angulo, Rosado-Paredes, Flores-Flores, Beaty and Gubler1999; Pollitt et al., Reference Pollitt, Bram, Blanford, Jones and Read2015). For example, works on mosquitoes infected by Zika or Chikungunya virus have found a slight effect on the transmission rate of both viral species (Rückert et al., Reference Rückert, Weger-Lucarelli, Garcia-Luna, Young, Byas, Murrieta, Fauver and Ebel2017). A much-less studied case where multiple infections could prove useful is that of triatomine bugs and the parasite that they carry, Trypanosoma cruzi, the causative agent of Chagas disease. Trypanosoma cruzi infection reduces bug survival (e.g. Peterson et al., Reference Peterson, Graham, Elliott, Dobson and Chávez2016; Cordero-Montoya et al., Reference Cordero-Montoya, Flores-Villegas, Salazar-Schettino, Vences-Blanco, Rocha-Ortega, Gutiérrez-Cabrera, Rojas-Ortega and Córdoba-Aguilar2019; reviewed by de Oliveira et al., Reference De Oliveira, Alevi, Imperador, Madeira and de Azeredo-Oliveira2018). This cost could lead the insect to enhance its immune response (de Fuentes-Vicente et al., Reference De Fuentes-Vicente, Cabrera-Bravo, Enríquez-Vara, Bucio-Torres, Gutiérrez-Cabrera, Vidal-López, Martínez-Ibarra, Salazar-Schettino and Córdoba-Aguilar2016), removing up to 80% of the parasites within hours after ingesting infected blood (Elliot et al., Reference Elliot, Rodrigues, Lorenzo, Martins-Filho and Guarneri2015; Hinestroza et al., Reference Hinestroza, Ortiz and Molina2016).
Several studies have been conducted with different pathogen agents for the biological control of bugs (de Fuentes-Vicente et al., Reference De Fuentes-Vicente, Gutiérrez-Cabrera, Flores-Villegas, Lowenberger, Benelli, Salazar-Schettino and Cordoba-Aguilar2018). One of such pathogens is the fungus Metarhizium anisopliae, which penetrates the insect cuticle, suppressing its immune response and causing a high mortality (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016, Reference Flores-Villegas, Cabrera-Bravo, Pérez-Torres, Córdoba-Aguilar, Salazar-Schettino, Hernández-Velázquez and Toriello2018). Only one study has considered the combined effects of co-infection by T. cruzi and other pathogens used as control agents (Garcia et al., Reference Garcia, de Paula Rocha, Moreira, Rocha, Guarneri and Elliot2016). Paradoxically, that work reported higher survival rates when the triatomine Rhodnius prolixus (Hemiptera: Reduviidae) was coinfected by the fungus Beauveria bassiana and T. cruzi than those caused by the fungal infection alone (Garcia et al., Reference Garcia, de Paula Rocha, Moreira, Rocha, Guarneri and Elliot2016).
Herein, survival and immune response of the triatomine bug Meccus pallidipennis (Hemiptera: Reduviidae) after a multiple infection by M. anisopliae and T. cruzi was studied. This bug is one of the most highly widespread in Mexico, and its occurrence out and inside houses makes it a key component for Chagasic control measures (de Fuentes-Vicente et al., Reference De Fuentes-Vicente, Gutiérrez-Cabrera, Flores-Villegas, Lowenberger, Benelli, Salazar-Schettino and Cordoba-Aguilar2018). Its longevity, from egg to adult and when free of T. cruzi infection, is approximately 171 days (at 28°C temperature and 60% relative humidity) (Bautista et al., Reference Bautista, Rojas, De Haro, Bucio and Salazar-Schettino2001) depending on its geographical origin (Martínez-Ibarra et al., Reference Martínez-Ibarra, García-Benavidez, Vargas-Llamas, Bustos-Saldaña and Montañez-Valdes2012). Eighty days has been reported for the length of the fifth stage nymph (Martínez-Ibarra and Katthain-Duchateau, Reference Martínez-Ibarra and Katthain-Duchateau1999), which is the stage that we used in our present study. Notice, however, that such longevity or life periods are shorter when a T. cruzi infection occurs (Cordero-Montoya et al., Reference Cordero-Montoya, Flores-Villegas, Salazar-Schettino, Vences-Blanco, Rocha-Ortega, Gutiérrez-Cabrera, Rojas-Ortega and Córdoba-Aguilar2019).
Considering that the immune response of the insect is diminished by both M. anisopliae (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016, Reference Flores-Villegas, Cabrera-Bravo, Pérez-Torres, Córdoba-Aguilar, Salazar-Schettino, Hernández-Velázquez and Toriello2018) but enhanced by T. cruzi infection (De Fuentes-Vicente et al., Reference De Fuentes-Vicente, Cabrera-Bravo, Enríquez-Vara, Bucio-Torres, Gutiérrez-Cabrera, Vidal-López, Martínez-Ibarra, Salazar-Schettino and Córdoba-Aguilar2016), it is not clear what the net effect of a multiple infection will be on M. pallidipennis immune response. However, given the possible synergistic effect of both pathogens, our working hypothesis is that a multiple infection will lead to a decrease in bug survival with respect to single infections by M. anisopliae or T. cruzi. Phenoloxidase (PO) activity, a key part of the immune defense in insects (González-Santoyo and Córdoba-Aguilar, Reference González-Santoyo and Córdoba-Aguilar2012), will be used as a proxy of the bug immune response. Notice that the mechanisms of infection of T. cruzi and M. anisopliae are likely different. On one hand, T. cruzi develops its life cycle in the intestine where PO may not necessarily participate directly, yet it is expressed as a bug's systemic response (Favila-Ruiz et al., Reference Favila-Ruiz, Jiménez-Cortés, Córdoba-Aguilar, Salazar-Schettino, Gutiérrez-Cabrera, Pérez-Torres, De Fuentes-Vicente, Vences-Blanco, Bucio-Torres, Flores-Villegas and Cabrera-Bravo2018). On the other hand, M. anisopliae triggers melanization processes, where PO is involved (reviewed by Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016). Thus, for the parasite and fungus, PO can be used as a reliable way to assess pathogen attacks as a bug's systemic and defense response, respectively (González-Santoyo and Córdoba-Aguilar, Reference González-Santoyo and Córdoba-Aguilar2012).
Materials and methods
Insects
Meccus pallidipennis specimens were obtained from a colony maintained at the insectarium of the Laboratory of Parasite Biology, Department of Microbiology and Parasitology, Medicine Faculty, Universidad Nacional Autónoma de México. This colony was established in 1998 from insects collected in Oaxtepec, Morelos, Mexico (18°54′23″N, 98°58′13″W). All animals were maintained under controlled conditions, 60% RH, 28°C, and 12/12 h light–darkness cycles. Triatomine specimens were given the treatments described below (for survival and immune response assessment) one day after moulting to the fifth-instar nymphal stage. For each of the four experimental groups described below (either to assess survival or immune response), ten insects were used for each trial which were repeated three times, resulting in 30 insects in total per treatment. All trials were run simultaneously.
Trypanosoma cruzi
The ITRI/MX/12/MOR T. cruzi strain (named according to WHO, 2002) was used in this study. This strain was isolated from a M. pallidipennis specimen captured in 2012 in Oaxtepec, Morelos. This isolate is maintained in the repository of the Laboratory of Parasite Biology, DPP-FM-UNAM. It was typified as TcI and is maintained by successive passages to CD-1 mice (Cordero-Montoya et al., Reference Cordero-Montoya, Flores-Villegas, Salazar-Schettino, Vences-Blanco, Rocha-Ortega, Gutiérrez-Cabrera, Rojas-Ortega and Córdoba-Aguilar2019).
Metarhizium anisopliae
Monosporic cultures from the EH-473/4 M. anisopliae isolate were used. This strain was isolated in 1994 from a specimen of Aeneolamia sp. (Hemiptera: Cercopidae) collected in a sugarcane culture field in San Luis Potosí, Mexico. The isolate is maintained in the strain repository of the Laboratory of Basic Mycology, DPP-FM-UNAM, and registered in the World Federation of Culture Collections (WFCC) as BMFM-UNAM 834 (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016). The isolate was previously genotypically and phenotypically characterized in terms of its virulence to insects and innocuity to mammals (Toriello, Reference Toriello2003; Toriello et al., Reference Toriello, Pérez-Torres, Burciaga-Díaz, Navarro-Barranco, Pérez-Mejía, Lorenzana-Jiménez and Mier2006, Reference Toriello, Montoya-Sansón, Zavala-Ramírez, Navarro-Barranco, Basilio-Hernández, Hernández-Velázquez and Mier2008). Also, its DNA sequence was identified as indicated by Brunner-Mendoza et al. (2018).
Experimental groups
Infection by T. cruzi
Insects were allowed to feed on CD-1 mice weighing 20–25 g, previously inoculated with 107 trypomastigotes of the ITRI/MX/12/MOR strain. At the moment of feeding, the parasite was at the exponential growth stage while in the mice. Bug specimens were allowed to feed for 20 min in the dark.
Infection by M. anisopliae
Conidia were suspended in 3 ml of 0.5% Tween 80. After adjusting concentration to 107 conidia ml−1, the suspension was poured into Petri dishes containing PDA. Conidium germination status was determined by observation under a 40× objective. A conidium was regarded as germinated when the germ tube length was greater than or equal to conidium width. Two hundred germinated conidia were counted and placed in each Petri dish. Thirty microliters of a suspension containing 107 microconidia ml−1 were applied to experimental insects. Each insect was individually placed in a plastic Petri dish on sterile filter paper and incubated at 28°C, 80% RH, under a 12/12 h light–darkness cycle in a precision incubator 818 (Thermo Scientific, Waltham, MA, USA).
Coinfection by T. cruzi and M. anisopliae
Insects were infected by T. cruzi (as described above) and, 20 min later, were infected by M. anisopliae as indicated above.
Infection-free, control group
Animals were allowed to feed on uninfected CD-1 mice weighing 20–25 g. Bug specimens were allowed to feed for 20 min in the dark.
Confirmation of infection
For those two groups where parasite infection took place (a and c), the presence of parasites was checked 15 days after infection for all animals. For this, rectal contents were obtained by gently pressing the distal abdominal area. Then, the contents were placed in a polypropilene microcentrifuge Eppendorf tube in 20 µl (0.9%) of isotonic saline solution. The sample was homogenized for 5 s. A 1:10 dilution was taken to count parasites in a Neubauer chamber at 40× magnification using a Stemi 2000 C stereoscopic microscope (Carl Zeiss, Oberkochen, Germany). For fungal infection in the two groups where this took place (b and c), we took fungal samples from all insects. This was done by taking some sporulating fungi emerging from the insect cuticle, using an inoculating loop under a flow hood. Fungal samples were cultivated in PDA cultures, to corroborate the micro- and macroscopic characteristics of M. anisopliae. We additionally took some imprints. For this, the insect cuticle surface was cleaned using 70% ethanol and 40% sodium hypochlorite to remove potential contaminants. Subsequently, a longitudinal (from head to posterior end) cut was gently made using scissors and the cuticle was removed. A drop of blue cotton was then placed on the internal area of the cuticle, and this structure was placed on a slide, covered with a coverslip and sealed with nail varnish to be observed under a microscope at 40× (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016). From all insects used, we observed 90% of sporulation on dead insects.
Survival
The above four experimental groups were observed daily for 60 days. To assess whether they were alive/dead, each animal was gently manipulated with dissecting forceps when no signs of life were observed. Those bugs that failed to move for 5 min after manipulation were regarded as dead. The experiment stopped at day 60, which was when all infected specimens were dead while all control specimens were still alive (see Results).
Hemolymph extraction and anterior midgut dissection
We had new experimental animals according to the groups indicated above (also ten animals per group repeated three times having 30 animals in total). Hemolymph was extracted on day 5 post-infection, by puncturing with a 27G-gauge needle-equipped syringe in the joint between the third limb and the thorax of the insect, right in the membrane that separates the coxa from the trochanter. The abdomen was slightly pressed near the zone to induce the flux of hemolymph through the wound (Favila-Ruiz et al., Reference Favila-Ruiz, Jiménez-Cortés, Córdoba-Aguilar, Salazar-Schettino, Gutiérrez-Cabrera, Pérez-Torres, De Fuentes-Vicente, Vences-Blanco, Bucio-Torres, Flores-Villegas and Cabrera-Bravo2018). The sample was collected with a 10 µl micropipette, placed in an Eppendorf tube containing 20 µl of PBS 1X, pH 7.2, and mixed with PBS to a final 1:20 dilution. The tubes were kept on ice during sample processing and stored at 4°C until used.
The anterior midgut (AMG) of insects of each experimental group was dissected on day 5 post-infection. Each specimen was placed in a Petri dish on ice, and dissected under a Stemi 2000 C stereoscopic microscope (Carl Zeiss). The abdomen of each insect was cleaned with a 70% alcohol-impregnated cotton wad, removing the limbs to prevent their movement. The fat body and Malpighian tubules were removed with entomological forceps under a stereoscopic microscope, and then the AMG was dissected and placed in a tube with 200 µl of PBS 1X, pH 7. The AMG was homogenized in PBS with a pestle and the suspension was then centrifuged at 9168 × g for 10 min at 4°C. The supernatant was diluted 1:10 with PBS (Favila-Ruiz et al., Reference Favila-Ruiz, Jiménez-Cortés, Córdoba-Aguilar, Salazar-Schettino, Gutiérrez-Cabrera, Pérez-Torres, De Fuentes-Vicente, Vences-Blanco, Bucio-Torres, Flores-Villegas and Cabrera-Bravo2018)
Total protein quantification
Protein content in hemolymph and MGA samples was quantified and standardized by the Pierce method, using the commercial BCA Kit (Thermo Fisher Scientific, Rockford, IL, USA). Ten microliters of each sample (hemolymph or MGA) were placed in a 96-well microplate; 40 µl of PBS 1X, pH 7.2, and 150 µl of Pierce reagent mixture were added to each well. Albumin 2 mg ml−1 was used to plot a standard curve. The microplate was covered with aluminum foil and incubated at 37°C for 30 min. The absorbance was determined in an ELX 800 ELISA reader (Biotek Instruments, Winooski, VT, USA) at 562 nm. Protein content was adjusted to 10 µg of protein to measure PO activity (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016).
Phenoloxidase activity in hemolymph and the anterior midgut
PO activity was quantified by spectrophotometry through the catalytic conversion of L-dopa (colorless) into dopachrome (reddish brown), using a molar extinction coefficient of 3.715 M−1 cm−1 at 490 nm (Daquinag et al., Reference Daquinag, Nakamura, Takao, Shimonishi and Tsukamoto1995). Enzymatic activity was expressed as enzymatic units (U), where 1 U is the amount of enzyme that produces 1 µmol of dopachrome per min (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016; Favila-Ruiz et al., Reference Favila-Ruiz, Jiménez-Cortés, Córdoba-Aguilar, Salazar-Schettino, Gutiérrez-Cabrera, Pérez-Torres, De Fuentes-Vicente, Vences-Blanco, Bucio-Torres, Flores-Villegas and Cabrera-Bravo2018).
PO in hemolymph
Ten micrograms of protein from each sample in 100 µl of PBS were placed in a 96-well microplate; 100 µl of substrate (L-dopa, 4 mg ml−1) were added, and the plate was incubated at 37°C for 20 min in the dark. Absorbance was measured in an ELX 800 ELISA reader (Biotek Instruments) at 490 nm every 5 min for 1 h. A blank solution was included, containing 100 µl of PBS plus 100 µl of L-dopa.
PO in AMG
Ten microliters of PBS were placed in a 96-well microplate, plus 25 µl of homogenized AMG sample and 25 µl of substrate (L-dopa, 4 mg ml−1). The plate was incubated at 37°C for 3 h in the dark. Absorbance was measured in an ELX 800 ELISA reader (Biotek Instruments) at 490 nm every 5 min for 1 h. A blank solution was included, containing 35 µl of PBS plus 25 µl of L-dopa. The assay was performed in duplicate.
Statistical analysis
A Kaplan–Meier analysis was performed to obtain a survival curve. Mean survival and risk index (at 95% of confidence) were analyzed with a log-rank test (Mantel–Cox). Several transformations were used to analyze PO activity in hemolymph, but none yielded variance homogeneity or normality of data. Thus, a Kruskal–Wallis test was used for this variable. A Bonferroni-corrected Dunn's test was used as a post-hoc test for multiple comparisons, thus reducing the probability of false negatives. We used a square-root transformation for PO activity in AMG. This allowed us to use an ANOVA having PO activity as the response variable and the experimental groups as factors. A Tukey test was used as a post-hoc test. For statistical analyses, we used SPSS statistics Vers. 22.
Results
Confirmation of infection
On groups (a) infection by T. cruzi and (c) coinfection by T. cruzi and M. anisopliae, the rectal contents had an average of 208,936 parasites ml−1 ± 103,456 (STD). A successful fungal infection was confirmed for groups (b) Infection by M. anisopliae and (c) coinfection by T. cruzi and M. anisopliae by the presence of hyphae, mycelium, and sporulation.
Survival
Differences in survival were observed among experimental groups (fig. 1). The lowest survival rates were observed in the M. anisopliae group (mean survival time (MST) = day 7 (SE 0.56); hazard ratio (HR) (95% CI) = 0.01237–0.1296; P = <0.0001, X 2 = 28.83, df = 1), followed by T. cruzi + M. anisopliae (MST = day 12 (SE 0.87); HR (95% CI) = 0.01507–0.1398; P = <0.0001, X 2 = 24.86, df = 1), T. cruzi (MST = day 47 (SE 3.07); HR (95% CI) = 0.08575–0.7471; P = 0.0440, X 2 = 3.517, df = 1), and control subjects (MST = not defined (SE 4.32); HR (95% CI) = not defined; P = not defined).
Cumulative survival of Meccus pallidipennis nymphs after four treatments (infection-free, T. cruzi, M. anisopliae, and T. cruzi + M. anisopliae; n = 30 for each group).
PO in hemolymph and AMG
Significant differences in PO activity in hemolymph were observed. The highest value was observed in the control group, followed by T. cruzi, T. cruzi + M. anisopliae, and M. anisopliae (Kruskal–Wallis test = 91.82, df = 3, P < 0.0001; fig. 2a). When intergroup comparisons were performed, significant differences emerged between control vs. M. anisopliae, control vs. T. cruzi + M. anisopliae, T. cruzi vs. M. anisopliae, and T. cruzi vs. T. cruzi + M. anisopliae (all P < 0.001; fig. 2a).
Phenoloxidase activity (mean ± standard error) for (a) hemolymph and (b) anterior midgut from Meccus pallidipennis nymphs after four treatments (infection-free, T. cruzi, M. anisopliae, and T. cruzi + M. anisopliae; n = 30 for each group).
Analysis of variance yielded differences in the main model for PO activity in AMG (F 1,2 = 76.87, P < 0.0001; fig. 2b). Again, significant differences emerged between control vs. M. anisopliae, control vs. T. cruzi + M. anisopliae, T. cruzi vs. M. anisopliae, and T. cruzi vs. T. cruzi + M. anisopliae (intergroup Tukey test for all, P < 0.0001; fig. 2b).
Discussion
Our results indicated that the fungus M. anisopliae kills the triatomine bug M. pallidipennis faster when acting alone than when acting in combination with T. cruzi, or when the protozoan is the only pathogen agent. Similar survival results when using one vs. multiple pathogens have been already observed in other insects, for example, Drosophila melanogaster (e.g. McClure et al., Reference McClure, Zhong, Hunt, Chapman, Hill and Priest2014). Metarhizium anisopliae is widely used in the biological control of pests and disease-transmitting insects (reviewed by Brunner-Mendoza et al., Reference Brunner-Mendoza, Reyes-Montes, Moonjely, Bidochka and Toriello2019). In fact, it has been used to control M. pallidipennis, leading to insect death within a few days after infection (Flores-Villegas et al., Reference Flores-Villegas, Cabrera-Bravo, Toriello, Bucio-Torres, Salazar-Schettino and Córdoba-Aguilar2016, Reference Flores-Villegas, Cabrera-Bravo, Pérez-Torres, Córdoba-Aguilar, Salazar-Schettino, Hernández-Velázquez and Toriello2018). Why did the combined effect of the fungus with T. cruzi fail to reduce bug survival more than the fungus alone? Since T. cruzi depends on the bug's survival for the transmission of the former, it may be more favorable for the vector to invest in damage control and repair (i.e. a tolerance strategy) than to invest in resistance. This could explain why we did not detect an upregulated immune response toward T. cruzi infection, and it may also explain the significantly longer lifespan in the presence of M. anisopliae.
Notice that similar results have been already documented in triatomines. For instance, Garcia et al. (Reference Garcia, de Paula Rocha, Moreira, Rocha, Guarneri and Elliot2016) found lower survival rates in the triatomine R. prolixus when infected by the fungus B. bassiana alone than when coinfected with T. cruzi and B. bassiana. The authors of that study suggested the existence of a protective effect by T. cruzi against fungal infections. Interestingly, multiple infections by T. cruzi and T. rangeli resulted in higher survival rates than single T. cruzi infections (Peterson et al., Reference Peterson, Graham, Elliott, Dobson and Chávez2016), supporting the hypothesis of a protective effect by T. cruzi against possible competitors. The authors suggested that the microbiota interacting with T. cruzi in the triatomine gut could mediate this protective effect by releasing antimicrobial compounds against the fungus (Garcia et al., Reference Garcia, de Paula Rocha, Moreira, Rocha, Guarneri and Elliot2016). This would imply some degree of bug manipulation by the parasite to improve its response against the fungus. In fact, bug manipulation by T. cruzi has been documented (reviewed by de Oliveira et al., Reference De Oliveira, Alevi, Imperador, Madeira and de Azeredo-Oliveira2018) via, for example, a decrease in the lapse between feeding and defecation (Botto-Mahan et al., Reference Botto-Mahan, Cattan and Medel2006) and in the motion activity (Marliére et al., Reference Marliére, Latorre-Estivalis, Lorenzo, Carrasco, Alves-Silva, de Rodrigues, de Ferreira, de Lara, Lowenberger and Guarneri2015). Decreasing the feeding–defecation lapse favors a rapid deposition of the parasite on the skin of a vertebrate host, while the reduction in bug motion activity may favor a longer interaction with the vertebrate. In general, these behavioral alterations would result in a more rapid and effective passage of T. cruzi to a vertebrate host. With respect to our results and those reported by Garcia et al. (Reference Garcia, de Paula Rocha, Moreira, Rocha, Guarneri and Elliot2016) manipulating the bug immune defense would increase its lifespan, which ultimately would favor the passage of T. cruzi to a vertebrate host. While these speculations have been proposed before for vector insects (Hurd, Reference Hurd2003), further experiments are required to demonstrate that the possible bug manipulation by the parasite actually results in genetic and physiological changes favoring the successful passage of the parasite to a new host (Herbison et al., Reference Herbison, Lagrue and Poulin2018).
As expected, PO activity levels in hemolymph and the AMG in the group coinfected by T. cruzi and the fungus were lower than in the control and T. cruzi-infected groups. Considering the possible bug manipulation discussed above, if a protective effect by T. cruzi exists, it does not seem to be mediated by the bug immune system. A possible explanation could be that the bug immune response is being reduced to save energetic resources and extend its life. This survival strategy has been observed in other animals; when the risk of death is imminent, the animal allocates most resources to extend its life as much as possible, even at the cost of drastically reducing its immune defense (reviewed by Schmid-Hempel, Reference Schmid-Hempel2011). This strategy may favor the bug and the parasite as well, allowing the former to live longer and probably to reach the reproductive age, while giving more time to the latter to pass to the next obligate host. On the other hand, a decrease in the bug immune response caused by single fungal and parasitic infections cannot be ruled out. This immunosuppression could also be a strategy to increase the reproductive capacity of the bug. However, T. cruzi infection is known to reduce the triatomine reproductive capacity (Fellet et al., Reference Fellet, Lorenzo, Elliot, Carrasco and Guarneri2014; Cordero-Montoya et al., Reference Cordero-Montoya, Flores-Villegas, Salazar-Schettino, Vences-Blanco, Rocha-Ortega, Gutiérrez-Cabrera, Rojas-Ortega and Córdoba-Aguilar2019), possibly as a cost of the consumption of energetic resources by the parasite.
Although we used PO as a systemic and immune defense response for the parasite and fungus, respectively, other immune components can be measured. Triatomines use two main signaling pathways: IMD and Toll. IMD is activated in the presence of lipopolysaccharides in Gram-negative bacteria (Ribeiro et al., Reference Ribeiro, Genta, Sorgine, Logullo, Mesquita, Paiva-Silva, Majerowicz, Medeiros, Koerich, Terra, Ferreira, Pimentel, Bisch, Leite, Diniz, Lídio da, Junior, Da Silva, Araujo, Caroline, Gandara, Brosson, Salmon, Bousbata, González-Caballero, Silber, Alves-Bezerra, Gondim, Silva-Neto, Atella, Araujo, Dias, Polycarpo, Vionette-Amaral, Fampa, Melo, Tanaka, Balczun, Oliveira, Gonçalves, Lazoski, Rivera-Pomar, Diambra, Schaub, Garcia, Azambuja, Braz and Oliveira2014; Vieira et al., Reference Vieira, Waniek, Mattos, Castro, Mello, Ratcliffe, Garcia and Azambuja2014), so that the first response is AMPs and defensins (reviewed by Flores-Villegas et al., Reference Flores-Villegas, Salazar-Schettino, Córdoba-Aguilar, Gutiérrez-Cabrera, Rojas-Wastavino, Bucio-Torres and Cabrera-Bravo2015). The Toll pathway has been reported to be induced by B-1,3-glucans present in fungi and in Gram-positive bacteria, triggering processes of melanization, coagulation, phagocytosis, and hemocyte encapsulation (Ribeiro et al., Reference Ribeiro, Genta, Sorgine, Logullo, Mesquita, Paiva-Silva, Majerowicz, Medeiros, Koerich, Terra, Ferreira, Pimentel, Bisch, Leite, Diniz, Lídio da, Junior, Da Silva, Araujo, Caroline, Gandara, Brosson, Salmon, Bousbata, González-Caballero, Silber, Alves-Bezerra, Gondim, Silva-Neto, Atella, Araujo, Dias, Polycarpo, Vionette-Amaral, Fampa, Melo, Tanaka, Balczun, Oliveira, Gonçalves, Lazoski, Rivera-Pomar, Diambra, Schaub, Garcia, Azambuja, Braz and Oliveira2014; reviewed by Flores-Villegas et al., Reference Flores-Villegas, Salazar-Schettino, Córdoba-Aguilar, Gutiérrez-Cabrera, Rojas-Wastavino, Bucio-Torres and Cabrera-Bravo2015), where the enzyme PO is involved (Azambuja et al., Reference Azambuja, Feder, Mello, Gomes and Garcia1999). Also, free radicals such as nitric oxide have proven to be a defense line with Trypanosoma rangeli in R. prolixus (Whitten et al., Reference Whitten, Mello, Gomes, Nigam, Azambuja, Garcia and Ratcliffe2001, Reference Whitten, Sun, Tew, Schaub, Soukou, Nappi and Ratcliffe2007). In this fashion, apart from PO other immune components to use to assess T. cruzi and fungal infections by the bug are AMPs, defensins, melanization, coagulation, phagocytosis, and nitric oxide.
Finally, our study has implications for the use of entomopathogenic fungi as biological control agents. Considering our results and those reported by Garcia et al. (Reference Garcia, de Paula Rocha, Moreira, Rocha, Guarneri and Elliot2016), those bugs infected by T. cruzi would live longer than infection-free insects when treated with the fungus. Our results indicate that the fungus alone kills 100% of the triatomine population in 10 days, while the combined effect of the fungus and the parasite kills 100% of triatomine population in 15 days. If this 5-day difference means that fungus-treated, T. cruzi-infected bugs may still have a significantly high vectorial activity, the fungal strategy should be accompanied by other control methods.
Acknowledgements
This work was funded by DGAPA-PAPIIT, grant No. IN216118 and IN227816. Thanks to Dr Conchita Toriello Nájera for providing the fungus, as well as Hortensia Navarro Barranco for her advice to handle the fungus. Thanks to Mauro Omar Vences Blanco for providing both M. pallidipennis and advice on their maintenance.
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical standards
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. All procedures performed in studies involving animals were in accordance with the ethical standards of the institution at which the studies were conducted.
